[Bug driver/47249] [4.4/4.5/4.6/4.7 Regression] ICE in common_handle_option, at opts.c:1695 with unknown option passed to cc1

[jakub at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=47249

--- Comment #5 from Jakub Jelinek  2012-01-24 
13:18:16 UTC ---
Author: jakub
Date: Tue Jan 24 13:18:08 2012
New Revision: 183474

URL: http://gcc.gnu.org/viewcvs?root=gcc&view=rev&rev=183474
Log:
PR driver/47249
* common.opt (-pie, -shared, pie, shared): Change from Common to
Driver.
* gcc.c (display_help): Display help for -pie and -shared.

Modified:
trunk/gcc/ChangeLog
trunk/gcc/common.opt
trunk/gcc/gcc.c

[Bug driver/47249] [4.4/4.5/4.6/4.7 Regression] ICE in common_handle_option, at opts.c:1695 with unknown option passed to cc1

[rguenth at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=47249

Richard Guenther  changed:

   What|Removed |Added

   Priority|P3  |P2

--- Comment #4 from Richard Guenther  2012-01-19 
12:17:17 UTC ---
I suppose we should avoid the ICE for releases by simply turning the gcc_assert
into a gcc_checking_assert ...

[Bug driver/47249] [4.4/4.5/4.6/4.7 Regression] ICE in common_handle_option, at opts.c:1695 with unknown option passed to cc1

[jakub at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=47249

--- Comment #3 from Jakub Jelinek  2012-01-16 
13:51:32 UTC ---
Created attachment 26340
  --> http://gcc.gnu.org/bugzilla/attachment.cgi?id=26340
gcc47-pr47249-2.patch

Or this one (or something else)?  Having driver display_help use *.opt stuff is
probably beyond stage4 material.

[Bug driver/47249] [4.4/4.5/4.6/4.7 Regression] ICE in common_handle_option, at opts.c:1695 with unknown option passed to cc1

[jakub at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=47249

--- Comment #2 from Jakub Jelinek  2012-01-16 
13:50:11 UTC ---
Created attachment 26339
  --> http://gcc.gnu.org/bugzilla/attachment.cgi?id=26339
gcc47-pr47249.patch

Joseph, for 4.7, do you prefer this version